Background estimation in the search for single production of vector-like quarks decaying into $Wb$ in $pp$ collisions using a data-driven method at $\sqrt{s} = 13$ TeV with the ATLAS detector

TL;DR

This paper presents a data-driven ABCD method to estimate QCD multijet background in a search for single vector-like quark production decaying to Wb at 13 TeV with the ATLAS detector, addressing modeling challenges.

Contribution

It introduces a novel ABCD background estimation technique that accounts for correlations among control regions in a complex hadronic decay channel.

Findings

The ABCD method effectively estimates QCD multijet background.

Correlation factors improve the accuracy of background estimation.

The method's performance varies with different jet collections and taggers.

Abstract

A search for singly produced vector-like $T/Y$ quark is performed in proton-proton collision data at a centre-of-mass energy of $13$ TeV corresponding to an integrated luminosity of $139.1 fb^{-1}$, recorded with the ATLAS detector at the LHC from 2015-18. The analysis focused on the hadronic channel of $T/Y \rightarrow Wb$ decay mode. The hadronic channel is not a clean channel, unlike the leptonic channel, because of the contribution coming from the QCD processes, which accounts for a significant background. The modelling of QCD processes is challenging due to their theoretical prediction and experimental challenges. Due to this, it is observed that the Monte Carlo simulation of the QCD multijet is mismodelled in this analysis. In this thesis, a data-driven method called the ABCD method is presented to estimate the QCD multijet background. The method uses the background events from all the control regions to estimate the multijets in the signal region. However, a correlation among different control regions is observed, which is further accounted by introducing a correlation factor. The correlation factor is calculated by performing likelihood fit to the MC simulated backgrounds. Some additional studies, including uncertainty estimation on the method and the performance of the method with different jet collections and taggers, are also performed.